Radio tuning problem. Need ideas/guidance.

[StellarRat]

I'm pretty much a beginner in electronics, so please be patient with me. I've built a little 3 FET AM detector radio on my breadboard. I attached an op amp circuit to the output to boost the sound (not shown.) My tuning LC circuit consists of a loop antenna (L1) that I believe is around 240 uH and 420 pF NTE618 varactor w/pot as my tuner (this replaces C1 the variable mechanical capacitor. C2 to gate of Q1 OR gate of RF amp) Together all these parts work very well and I'm able to tune the entire AM from 540 kHz to 1490 kHz (possibly higher, but there are no stations that high in my area that are strong enough to detect.)

My problem is that when I add in an RF amp (shown lower right) between the LC tuner and my detector (RF out to gate of Q1) I lose much of the top of the AM band. The part that remains does have much better reception, so I know the RF amp is doing it's job. What could be causing my trouble and how can I fix it?

My current "theories" about what might be wrong are: 1. the RF amp is somehow throwing the inductance or capacitance of the LC off. I've tried compensating with capacitors in series and inductors in parallel to push the range "up", but even though I can get that to work I end up losing the bottom end of band and not gaining much on the top, so the tuning range is still much reduced from the non-amplified version. 2. the RF amp is somehow changing how the antenna works. 3. somehow the RF amp is interfering with functioning of the varactor tuner.

Anyway, any help or guidance would be appreciated. I have a multi-function meter, cheesy oscilloscope and a variable power supply available to test with.

I've attached a jpg with the relevant schematics of the various pieces. The top is the original detector circuit. The lower left is the varactor tuner and the lower right is the RF Amp I'm trying to use.

Note: There is only one LC tank circuit. It's just replicated in the schematics.

 

[MikeMl]

When you add the preamp, are there two sets of LC tanks in the chain?

Track-tuning two circuits simultaneously is a real trick...

 

[StellarRat]

When you add the preamp, are there two sets of LC tanks in the chain?

Track-tuning two circuits simultaneously is a real trick...

No, only one tank. It's just replicated in the drawings. Sorry about the confusion.

 

[MikeMl]

Then a merged schematic would help...

 

[StellarRat]

Then a merged schematic would help...

This is the best I can do with the tools I have.

 

[JimB]

One thing that is obviously missing from your modified circuit is a resistor from the gate of the Q1 FET to the 0v (gnd) line.
Add a 1M Ohm resistor as per my sketch.

There may be other problems with the layout of your circuit, but that is difficult to now without seeing a picture.

JimB

 

[audioguru]

I suspect that the circuit with the preamp oscillates at the high end of the AM broadcast band due to using a Mickey Mouse solderless breadboard with its intermittent contacts and lots of stray capacitance coupling between all the rows of contacts and the
messy jumper wires all over the place.

 

[StellarRat]

One thing that is obviously missing from your modified circuit is a resistor from the gate of the Q1 FET to the 0v (gnd) line.
Add a 1M Ohm resistor as per my sketch.

There may be other problems with the layout of your circuit, but that is difficult to now without seeing a picture.

JimB

View attachment 92799

Thank you! I'll try that. Now, showing my beginnerness, what does that do/why? Thanks! BTW, that greatly improved the selectivity and sensitivity of the radio, but it did not give me the upper end of the tuning range. I'm going to re-try the capacitor in series with the varactor and see if that fixes it.

 

[Willen]

Probably the resistor from Gate to 0V helps to discharge FET's Gate-Source junction capacitance, and probably it helps to swing gate and drain signal efficiently, isn't it JimB?

So sorry, I am not a crazy, poor cellphone network posted lot of duplicate posts, as below. I cannot delete them so edited.

 

[StellarRat]

One thing that is obviously missing from your modified circuit is a resistor from the gate of the Q1 FET to the 0v (gnd) line.
Add a 1M Ohm resistor as per my sketch.

There may be other problems with the layout of your circuit, but that is difficult to now without seeing a picture.

JimB

View attachment 92799

Thank you! I'll try that. Now, showing my beginnerness, what does that do/why? Thanks! BTW, that greatly improved the selectivity and sensitivity of the radio, but it did not give me the upper end of the tuning range. I'm going to re-try the capacitor in series with the varactor and see if that fixes it.

Tried the capacitors no luck, but it's still better than before. I think I'm missing about 300 - 500 kHz from the top of the band.

 

[audioguru]

The top of the radio band is missing because the capacitance across the coil is too high.

The original circuit uses the Jfet as a source follower so that the drain to gate capacitance is low.
But the preamp amplifies the drain to gate capacitance (Miller effect) which increases it and it is across the coil.

The varactor diode has a low capacitance when its voltage is high so it might need 12V instead of whatever your voltage is. What is your voltage? 6V from a dead 9V battery?
The tuning will change as the battery voltage runs down.

 

[JimB]

Probably the resistor from Gate to 0V helps to discharge FET's Gate-Source junction capacitance, and probably it helps to swing gate and drain signal efficiently, isn't it JimB?

No it isn't Willen !

It is to do with the biasing of the FET.
FETs like the MPF102 and 2N3819, for correct operation they should be biased so that the gate is negative with respect to the source.
In this respect they are just like thermionic valves where in normal operation the grid is negative with respect to the cathode.

Have a look at the circuit fragment which I have attached to this post, it is a simlified version of what I drew in post #6.
Conventional current is flowing through the FET from the +ve supply to the 0v line via the drain and the source.
As that current flows through R1 (yes there are two of them, take your pick!), a voltage is developed across the resistor as per the usual V = I.R Ohms law thing.
When we connect the gate to the 0v line, either by a high value resistor, or the low resistance of a coil, the gate is connected to the -ve end of the resistor and the source to the +ve end of the resisitor.
And so, we magically have the correct biasing for our FET, gate -ve W.R.T. source.
The term for this type of thing is "self biasing" or "source resistor biasing".

JimB

 

[StellarRat]

The top of the radio band is missing because the capacitance across the coil is too high.

The original circuit uses the Jfet as a source follower so that the drain to gate capacitance is low.
But the preamp amplifies the drain to gate capacitance (Miller effect) which increases it and it is across the coil.

The varactor diode has a low capacitance when its voltage is high so it might need 12V instead of whatever your voltage is. What is your voltage? 6V from a dead 9V battery?
The tuning will change as the battery voltage runs down.

I have a fresh 9v in there. I tried raising the voltage all the way to 15v and it made no difference. I did a test and I do believe your answer might be the correct one. I took out the varactor tuner and swapped in a "real" 365pf variable which should, in theory, get close to 0 pf when wide open and , lo and behold, I still wasn't getting the top end anymore. To me that indicates that the built in capacitance of the amp is high enough that I simply can't tune the top end, period. What I need is some kind of amp that doesn't have any capacitance (or very little.) Ideas? Schematics? BTW, one other thing I noticed is that there is a powerful station near the top that is partially swamping the higher stations when the amp is in use, but I don't that is the main problem, it's probably a side effect.

 

[Colin]

You may have too many turns on your coil.
Remove the cap and see what station you are getting.
Try a frame coil.
See my article on the Crystal Set:

<Mod edit: Colin, for the last time, linking to your personal site is considered self-promotion and/or spam and is not allowed here. If you have some information for the thread starter, post it here on the forum. This is your last warning.>

 

[audioguru]

I never built a lousy-sounding AM radio. My first radio was FM and sounded great. That was before they invented FM stereo so when stereo was invented I built a stereo multiplex decoder for it.

 

[Willen]

I made AM radio with tiny TO-92 AM radio chip- TA7642 and MK484. I am 25km far from station and reception of the radio was so nice and nice sound. No antenna needed for hundreds of km fat station. Just 1.5V cell needed and can power each earpiece.

Hi AG, did you used dedicated ICs or just transistors? Stuff designed by you? I think it's very years ago so...do you still have schematic of you radio?

 

[audioguru]

I made AM radio with tiny TO-92 AM radio chip- TA7642 and MK484. I am 25km far from station and reception of the radio was so nice and nice sound. No antenna needed for hundreds of km fat station. Just 1.5V cell needed and can power each earpiece.

The TA7642 has only one LC tuned circuit so either it will pickup and have interference from many AM radio stations or the Q or its LC is so high that it will be very narrow band and not produce medium and high audio frequencies.
It will not work in my city because there are many very powerful radio stations and a few might be picked up at the same time. It is good that it has Automatic Gain Control which might prevent it from being overloaded. My portable Sony AM/FM radio has never been used on AM but the FM sounds great. It is easily overloaded by all the local stations so it has a local/distant switch that attenuates the input when set to local but then it does not pickup distant stations.

Hi AG, did you used dedicated ICs or just transistors? Stuff designed by you? I think it's very years ago so...do you still have schematic of you radio?

In 1961 transistors were not sold anywhere yet so I used vacuum tubes to make the FM tuner, later I made the stereo multiplex adapter for it when FM stereo was invented then I made a vacuum tube stereo power amplifier. I also made many speaker systems that sound great.

 

[Willen]

In 1961 transistors were not sold anywhere yet so I used vacuum tubes to make the FM tuner.

As I heard FM Rx is complicated thing. I got that 'regenarative' type of very simple Transistor based FM receivers are totally not usable. Do you have any example diagram of the transistor FM Rx, based on same theory you used in your Vacuum FM Tunner?

 

[StellarRat]

OK, everyone, I've mostly solved my problem! By adding a few turns of wire interleaved in loop antenna windings at a 1 to 5 ratio and running those to the RF amplifier instead of the main antenna loop I've gotten my tuning range back and the selectively is back to normal too. However, I've lost quite a bit of incoming signal. I think this is because of the inefficiency of the power transfer between the main loop and the secondary loop. I also read that there is a better amplifier design that doesn't have much Miller Capacitance called a "Cascode".

That leaves me with three final questions:

1. How can I determine what the proper ratio of main to secondary windings should be (other than experimentation)? Right now I think I have too few secondary windings.

2. Can I achieve the same results with some type of transformer where the output from the main antenna feeds input to the transformer and output from the transformer is input to RF amplifier (so the secondary windings are no longer needed.) If so, how would I make/buy one?

3. Does anyone have a schematic for a cascode amplifier that will work on 9v and has a frequency range from about 550 kHz to 1700 kHz? I've seen several out on the web, but the all require 12 to 30 volts.

BTW, I know that it is far easier to just buy a radio "chip" or buy a radio and that FM sounds better, but this is a hobby for me and I'm trying to learn how radios work and the basics of electronics, so please excuse the impractically of what I'm doing. People still build steam engines for fun!

Thanks!

 

[audioguru]

this is a hobby for me and I'm trying to learn how radios work and the basics of electronics. People still build steam engines for fun!

The extremely simple "radio" is not teaching you anything about real super-heterodyne radio circuits and very little about the basics of electronics.
You can build a bicycle for fun but it won't teach you anything about cars.